The Study On The Typical Problems And Countermeasures Of C-E Machine Translated Versions

The development and widespread application of machine translation software has played an increasingly important role in translating activities,while the translation quality has always been criticized.It is still worth further exploring and analyzing whether the quality of machine translated versions is reliable or not.This thesis takes Google,Youdao,Baidu and Tencent translation software as research tools and selects special word phrases,concise expressions,as well as compound and complex sentences of different texts as the research corpus to evaluate the quality and reliability of machine translated versions based on the theoretical foundation of Skopos Theory and Text Typology.This thesis concludes that machine translation software has major advantages in rapid translation of common vocabulary,efficient extraction of rough information,and acceptable translation quality for informative texts.Furthermore,this thesis finds out four typical problems of MTS:(1)problems in translating fixed terms with Chinese characteristics;(2)problems in translating public signs;(3)problems in translating metaphors;(4)problems in presenting logic relations and text features.To efficiently solve the above four problems,the thesis puts forward four corresponding countermeasures to improve the quality of translated versions:(1)turning to authoritative official versions;(2)following conventional expressions and normative documents;(3)borrowing English expressions equivalent to the original metaphors;(4)analyzing inner logical relation and polishing the whole versions.